Simultaneous localization and map building method and medium for moving robot

• US 20090154791A1
• Filed: 05/23/2008
• Published: 06/18/2009
• Est. Priority Date: 12/17/2007
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Roh, Kyung Shik, Min, Seung Ki, Yoon, Sukjune
• Granted Patent: US 8,331,652 B2
• Time in Patent Office: Days
• US Class Current: 382/153
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59

Abstract

A simultaneous localization and map building (SLAM) method and medium for a moving robot is disclosed. The SLAM method includes extracting a horizontal line from an omni-directional image photographed at every position where the mobile robot reaches during a movement of the mobile robot, correcting the extracted horizontal line, to create a horizontal line image, and simultaneously executing a localization of the mobile robot and building a map for the mobile robot, using the created horizontal line image and a previously-created horizontal line image.

Claims

1. A simultaneous localization and map building method for a mobile robot, comprising:

extracting a horizontal line from an omni-directional image photographed everywhere the mobile robot is positioned during a movement of the mobile robot;

correcting the extracted horizontal line, to create a horizontal line image; and

simultaneously executing a localization of the mobile robot and building of a map for the mobile robot, using the created horizontal line image and a previously-created horizontal line image.

2. The simultaneous localization and map building method according to claim 1, wherein the horizontal line exists on the same plane as an epipolar plane of an omni-directional camera which photographs the omni-directional image.

3. A computer readable recording medium having recorded thereon a computer program for executing the simultaneous localization and map building method according to claim 1.

4. The simultaneous localization and map building method according to claim 1, wherein the correcting of the extracted horizontal line is executed using a Gaussian filter.

5. The simultaneous localization and map building method according to claim 4, wherein the created horizontal line image comprises at least two horizontal lines respectively created by filtering the extracted horizontal line and then arranged above and beneath the extracted horizontal line.

6. The simultaneous localization and map building method according to claim 1, wherein:

the extracted horizontal line comprises pixels, and has a feature point; and

the feature point is a dark pixel having a brightness less than the remaining pixels of the extracted horizontal line by one brightness level.

7. The simultaneous localization and map building method according to claim 6, wherein:

the created horizontal line image has the feature point;

coordinates of the feature point are derived using a relationship between the position of the mobile robot, and an angle defined between the mobile robot and the feature point.

8. The simultaneous localization and map building method according to claim 7 wherein the coordinates are determined by evaluating the following expression

$\tan $



α

= Y - y X - x 

[ tan 



α

1 - 1 tan 



α

2 - 1 tan 



α

3 - 1 tan 



α

4 - 1 ] ·

[ X 1 Y 1 ] = [ tan 



α

1 ·

x 1 - y 1 tan 



α

2 ·

x 2 - y 2 tan 



α

3 ·

x 3 - y 3 tan 



α

4 ·

x 4 - y 4 ] ,

wherein, “

a”

represents the angle defined between the mobile robot and the feature point, “

X, Y”

represents the coordinates of the feature point, and “

x, y”

represents the coordinates of the mobile robot.

9. The simultaneous localization and map building method according to claim 7, wherein the angle defined between the mobile robot and the feature point in the created horizontal line image is derived by matching the feature point in the created horizontal line image with a corresponding feature point of the previously-created horizontal line image.

10. The simultaneous localization and map building method according to claim 9, wherein the feature point matching is executed using a sum of absolute difference (SAD) method.

11. The simultaneous localization and map building method according to claim 7, wherein the simultaneous localization and map building for the mobile robot is executed by applying, to an extended Kalman filter, the position of the mobile robot, the position of the feature point, and the angle defined between the mobile robot and the feature point in the created horizontal line image.

12. A simultaneous localization and map building method for a mobile robot, comprising:

extracting a horizontal line from an omni-directional image photographed during a movement of the mobile robot;

matching the extracted horizontal line with a previously-extracted horizontal line, to derive an angle defined between the mobile robot and a feature point represented within the extracted horizontal line; and

finding a position of the feature point, using a variation in the angle depending on a moved position of the mobile robot.

13. The simultaneous localization and map building method according to claim 12, further comprising:

applying, to an extended Kalman filter, the moved position of the mobile robot, the position of the feature point, and the angle defined between the mobile robot and the feature point within the extracted horizontal line, after the position of the feature point is found, to build a map according to the moved position of the mobile robot.

14. The simultaneous localization and map building method according to claim 12, further comprising:

creating new horizontal lines arranged above and beneath the extracted horizontal line, after the extraction of the horizontal line, by filtering the extracted horizontal line by a Gaussian filter, and arranging the filtered horizontal line above or beneath the extracted horizontal line.

15. The simultaneous localization and map building method according to claim 14, wherein the newly-created horizontal lines are matched with previously-created horizontal lines, respectively, to derive an angle defined between the mobile robot and the feature point respectively represented on each of the newly-created horizontal lines.

16. The simultaneous localization and map building method according to claim 12, wherein the feature point is represented by a dark pixel included in pixels of the extracted horizontal line while having a brightness lower than the remaining pixels of the extracted horizontal line by one brightness level.

17. The simultaneous localization and map building method according to claim 12, wherein the horizontal line exists on the same plane as an epipolar plane of an omni-directional camera which photographs the omni-directional image.

18. A computer readable recording medium having recorded thereon a computer program for executing the simultaneous localization and map building method according to claim 12.

19. A robot comprising:

an image processor creating a horizontal line image from an omni-directional image;

a controller executing a simultaneous localization and map building operation (SLAM) using the created horizontal line image; and

20. The robot according to claim 19 further comprising a memory to store the created horizontal line image.

21. The robot according to claim 19 wherein the controller executing the SLAM comprises extracting a horizontal line existing on the same plane as an epipolar plane of an omni-directional camera from which the omni-directional image is taken.